undocumented windows messages

[Olivier Gilloire]

Check the "windows.h" header for this.

Olivier

Peter Lerner a écrit dans le message <3438CB95...@symbiose.com>...
>hello,
>
>subclassing an edit control yields some undocumented messages.
>
>they are at least neither documented in the VC5 online manuals
>nor in the header files, e.g. winuser.h ...
>
>++++++++++++++++++++++++++ snip +++++++++++++++++++++++++++++++
>LRESULT CALLBACK subclass_wndproc(HWND hwnd, UINT message,
> WPARAM wParam, LPARAM lParam)
>{
> ...
>
> switch(message)
> {
> case WM_PAINT: // 0x000F
> case WM_ERASEBKGND: // 0x0014
> case WM_NCCALCSIZE: // 0x0083
> case WM_NCPAINT: // 0x0085
> break;
> // etc ...
>
> // those 3 msgs codes are not documented ...
> case 0x0118: // 0x0118 ???
> case 0x0131: // 0x0131 ???
> case 0x0164: // 0x0164 ???
> break;
> }
>
> ...
>}
>++++++++++++++++++++++++++ snap +++++++++++++++++++++++++++++++
>
>does anyone know where i can find infos about those messages,
>or has found out itself ?
>
>mfg
>peter lerner

[Peter Lerner]

hello Olivier Gilloire,

og> Check the "windows.h" header for this.

"windows.h" is only a meta include file, that includes all other
*possibly* needed windows-include files. no constants are defined
in it.

i even checked the *whole* include directory against the strings
'0x0118', '0x0131' and '0x0164' which *usually* yields a result.

i did find nothing ... but that also means nothing, because the
code could be calculated from a base value. the usual include
files gave nothing to it.

[Peter Lerner]

hello,

[Philippe Mermoud]

The fact is: if they are not documented you shouldn't try to use them...
Unless you've got something very strange to do in your app.
--
Remove the '_' at the beginning of my email address to contact me.

Peter Lerner <pe...@symbiose.com> wrote in article
<3438CB95...@symbiose.com>...

[Thomas F. Mooney, III]

PMFJI,

WM_USER is defined as follows:

#define WM_USER 0x0400

The message numbers they're looking for are all less than 0x0400.

TFM3

Note: Spam-resistant e-mail address

Terry Butler <75567...@compuserve.com> wrote in article
<#IQlbu$08GA...@uppssnewspub04.moswest.msn.net>...
> Some controls use a lot a messages in the WM_USER range, which means you
may
> find your number as part of a "WM_USER+...", which obviously won't be the
> direct value.
>
> --
> Terry Butler
>

[Ken Thomas]

Myk Willis wrote:

> Peter Lerner wrote:
> [...]

> > subclassing an edit control yields some undocumented messages.

> [...]

> >
> > // those 3 msgs codes are not documented ...
> > case 0x0118: // 0x0118 ???
> > case 0x0131: // 0x0131 ???
> > case 0x0164: // 0x0164 ???

> [...]
>
> 0x118 is WM_SYSTIMER
>
> WM_SYSTIMER was discussed in one of matt pietrek's articles in
> MSJ a while back (you could probably find it on msdn).
>
> don't know about 0x131 or 0x164.
>
> mwillis

According to "Undocumented Windows" (my copy was printed June 1994):
0x0118 is WM_SYSTIMER (pg. 532)
0x0131 is WM_LBTRACKPOINT (pg. 526)

Admittedly, this is a Win16 book, but it should give you clues. I don't
see 0x0164 listed anywhere. I don't know if there is an equivalent
Win32 book, but at the rate the Win32 API keeps changing, there's no way
_any_ book could attempt to document all the undocumented stuff. :-)

Ken

[Joerg R. Schaible]

Hi Peter,

| subclassing an edit control yields some undocumented messages.
|

| they are at least neither documented in the VC5 online manuals

| nor in the header files, e.g. winuser.h ...

| // those 3 msgs codes are not documented ...
| case 0x0118: // 0x0118 ???
| case 0x0131: // 0x0131 ???
| case 0x0164: // 0x0164 ???

looking in Schulman's Undoc'd Windows you'll find two of the entries:

0x0118 is WM_SYSTIMER similar to WM_TIMER
0x0131 is WM_LBTRACKPOINT which is only mensioned in combination with a
normal list

But as Raymond said: Who bother?

--
Joerg

Greetings from the Black Forest, Germany
Remove 'nospam.' to reply